%0 Journal Article
%T 基于随机森林算法的机器学习分类研究综述<br>A Review of Machine Learning Classification Based on Random Forest Algorithm
%A 向进勇
%A 王振华
%A 邓芸芸
%J Artificial Intelligence and Robotics Research
%P 143-152
%@ 2326-3423
%D 2024
%I Hans Publishing
%R 10.12677/AIRR.2024.131016
%X 机器学习是实现人工智能的重要技术,随机森林算法是机器学习的代表算法之一。随机森林算法以简单、有效而闻名工业界和学术界，它是基于决策树的分类器，通过投票选择最优的分类树。随机森林算法有可变重要性度量、包外误差、近似度等优秀特性，因此随机森林被广泛的应用到分类算法中。目前，不仅在医学、农业、自然语言处理等领域被广泛提及，而且在垃圾信息分类、入侵检测、内容信息过滤、情感分析等方面都有广泛的应用。本文主要介绍了随机森林的构建过程以及随机森林的研究现状，主要从分类性能、应用领域以及分类效果加以介绍,分析随机森林算法优缺点以及研究人员对随机森林算法的改进，希望通过分析能够让初学随机森林算法的研究人员掌握随机森林的理论基础。<br />
Machine learning is an important technology to realize artificial intelligence, and random forest algorithm is one of the representative algorithms of machine learning. The random forest algorithm is well-known in industry and academia for its simplicity and effectiveness. It is a decision tree-based classifier that selects the optimal classification tree through voting. Random forest algorithm is widely used in classification algorithms because of its excellent characteristics such as variable importance measure, out-of-envelope error and approximation. At present, it is not only widely mentioned in medicine, agriculture, natural language processing and other fields, but also widely used in junk information classification, intrusion detection, content information filtering, sentiment analysis and other aspects. This paper mainly introduces the construction process of random forest and the research status of random forest, mainly from the classification performance, application field and classification effect, analyzes the advantages and disadvantages of random forest algorithm and the improvement of random forest algorithm by researchers, hoping that through analysis, researchers who have just learned random forest algorithm can master the theoretical basis of random forest.
%K 决策树，随机森林，机器学习<br>Decision Trees
%K Random Forests
%K Machine Learning
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=81974